Temporal neurofuzzy MAR algorithm for time series data in rule-based systems

Abstract

Introduces a neurofuzzy model for constructing a knowledge-base of temporal fuzzy rules obtained by a MAR (multivariate autoregressive) algorithm. The model described contains two main parts which are fuzzy-rule extraction and storage of them. The fuzzy rules are obtained from time series data using the MAR algorithm. Fuzzy linear functions with fuzzy number coefficients are used. The extracted rules are fed into the temporal fuzzy multilayer feedforward neural network.